基于信息熵重构经验模态分解神经网络的带式输送机故障诊断
Fault Diagnosis of Belt Conveyor Based on Information Entropy Reconstruction Empirical Mode Decomposition Neural Network
尤峰 1郭刚 1闫涛 1吴振彬 2卢海军2
作者信息
- 1. 中煤陕西榆林能源化工有限公司,陕西 榆林 719000
- 2. 中煤信息技术(北京)有限公司,北京 100020
- 折叠
摘要
提出了一种神经网络模型,用于带式输送机电机轴承的早期故障诊断.该模型采用经验模态分解与信息熵相结合的方法对信号进行重构,提取频域特征用于神经网络的训练,以实现高度精准的故障诊断.该方法能够有效应对噪声干扰,对早期故障信号具有良好的敏感性,其诊断准确率高达95.8%.
Abstract
Proposed a neural network model for the early fault detection of belt conveyor motor bearing.By integrating empirical mode decomposition and information entropy,the model reconstructs signals and extracts frequency domain features for neural network training to realize the very accurate fault diagnosis.This method can effectively mitigate noise interference and is highly sensitive to initial fault signals which achieves a diagnostic accuracy of 95.8%.
关键词
带式输送机/经验模态分解/神经网络/轴承故障诊断Key words
belt conveyor/empirical mode decomposition/neural network/bearing fault diagnosis引用本文复制引用
出版年
2025
NETL
NSTL
万方数据